In our previous post, we looked at a variety of examples of pronominal clitics in noun phrases and their phonological & syntactic distribution. Primarily, we saw that the clitic pronoun consistently attached itself to the most prominent phonological phrase possible while also maintaining grammaticality within its syntactic domain. Thus, we saw clitics appearing in the following positions, with  marking the clitic’s syntactic phrase boundary and () marking its phonological phrase boundary.
(2) (X [=clitic) Det N]
(3) (X [=clitic) Y NP]*
This distribution of pronominal clitics within the Noun Phrase, for the most part, parallels that of the clause. In the case of the clause, though, it is less clear as to whether the pronominal clitic should be viewed as forming a syntactic phrase with the verb.**
(5) XP (V=clitic)
(6) (XP=clitic) V
(7) (Comp=clitic) V
(8) (Particle=clitic) V
Now while in the vast majority of cases, the most prominent prosodic phrase is the lexical head of the clause’s focal constituent, there are a few places where this is not the case. And that is our focus here. Fundamentally, this fact is the reason why I have chosen a phonological or prosodic explanation for the pronominal clitic distribution rather than a solely pragmatic one. An entirely pragmatic account or an entirely syntactic account cannot explain all the data.*** Instead, pragmatics, syntax, and phonology must be view as pieces of a mutually constrained grammatical system.
Clitics Attaching to Topics
John 7:29 κἀκεῖνός με ἀπέστειλεν.
John 17:25 καὶ οὗτοι ἔγνωσαν ὅτι σύ με ἀπέστειλας·
John 21:22 σύ μοι ἀκολούθει.
2 Cor 12:11 Γέγονα ἄφρων, ὑμεῖς με ἠναγκάσατε.
Heb 2:13 ἰδοὺ ἐγὼ καὶ τὰ παιδία ἅ μοι ἔδωκεν ὁ θεός
Clitics Attaching to Verbs with Fronted Focal Constituents
Mark 6:25 θέλω ἵνα ἐξαυτῆς δῷς μοι ἐπὶ πίνακι τὴν κεφαλὴν Ἰωάννου τοῦ βαπτιστοῦ.
Mark 8:2 ὅτι ἤδη ἡμέραι τρεῖς προσμένουσίν μοι
Mark 10:36 ὁ δὲ εἶπεν αὐτοῖς· τί θέλετέ [με] ποιήσω ὑμῖν;
Mark 15:34 ὁ θεός μου ὁ θεός μου, εἰς τί ἐγκατέλιπές με;
John 8:47 διὰ τί ὑμεῖς οὐ πιστεύετέ μοι;
Clitics Attaching to Non-lexical Words
Matt 8:8 ἵνα μου ὑπὸ τὴν στέγην εἰσέλθῃς
Luke 10:40 εἰπὲ οὖν αὐτῇ ἵνα μοι συναντιλάβηται.
Rom 15:19 ὥστε με ἀπὸ Ἰερουσαλὴμ καὶ κύκλῳ μέχρι τοῦ Ἰλλυρικοῦ πεπληρωκέναι τὸ εὐαγγέλιον τοῦ Χριστοῦ
2 Cor 7:7 ὥστε με μᾶλλον χαρῆναι.
2 Cor 12:7 ἵνα με κολαφίζῃ, ἵνα μὴ ὑπεραίρωμαι.
Eph 6:19 ἵνα μοι δοθῇ λόγος ἐν ἀνοίξει τοῦ στόματός μου, ἐν παρρησίᾳ γνωρίσαι τὸ μυστήριον τοῦ εὐαγγελίου
Philm 17 εἰ οὖν με ἔχεις κοινωνόν, προσλαβοῦ αὐτὸν ὡς ἐμέ.
I have arranged the data here from least difficult to most difficult. At a basic level, the clitic attaching to a fronted Topic poses no challenge to my claims, but it is a rather rare occurrence. Topics tend not receive the same degree of prosodic stress that a focal constituent does. Likewise, the middle category is not necessarily a problem. I have already proposed an explanation for this phenomena elsewhere, but it deserves an extended treatment in order to determine where my explanation works beyond those limited examples. Finally, the last category of pronominal clitics attaching to non-lexical words is quite striking. And while I will attempt to understand and explain it with the data I have, a broader corpus will need to be examined for any sort of definitive conclusions.
* Structurally, I am using, broadly speaking, an X-Bar theoretic approach to phrase structure, following my work on discontinuous NP’s in Koine Greek, which is in turn based on the work of Devine and Stephens, though without any of their claims about movement. I do this for data presentation, only, as a means to syntactically distinguish the different syntactic positions that enclitics are appearing in. Thus, example (3) assumes the discontinuous, but configurational phrase structure proposed in an earlier post and is only annotated at the maximal phrase level. The full annotation would be: (X [=clitic) [V [NP]]]. It is also possible that example (2) should also be viewed as discontinuous to an extent with a structure more like (X [=clitic) [Det N]]. But it must be emphasized that nothing in my claims here is reliant upon the acceptance of X-Bar theory or any sort of generative grammar. My own approach to linguistic theory attempts to pull together the strengths of both formal and functional grammar and syntax – to the extent that frameworks such as LFG and HPSG provide a formal architecture that also allows for broader functional and cognitive approaches,
** That is to say, the question of whether Hellenistic Greek has a VP in its phrase structure should still be viewed as an open question – one that, personally, I have gone back and forth on several times now. It may very well be possible to draw a phrase structure parallel between the distribution of clause level pronominal clitics around the verb and the phrase level pronominal clitics relationship distribution around the noun, though whether prosodic phrasing should be viewed as evidence for syntactic phrasing is debatable.
*** Devine and Stephens (Discontinuous Syntax: Hyperbaton in Greek [Oxford, 2000], 120) in a sense do both. While they observe the tendency for pronominal clitics in the discontinuous Y1 position to attach to Focal constituents, they use this as evidence that hyperbaton must be an autonomously syntactic phenomena. Likewise, there are a number of scholars who have acknowledge the correlation between clitic pronominals and Focal constituents but because of the lack of uniformity refer to it only as a tendency. For example X, Finally, Ann Taylor’s article on genitive clitics within noun phrases rightly takes a prosodic/phonological approach, but says nothing at all about the relationship between prosody and pragmatic function.